【24h】

Chinese Texts Classification System

机译:中国文本分类系统

获取原文

摘要

In this article, we designed an automatic Chinese text classification system aiming to implement a system for classifying news texts. We propose two improved classification algorithms as two different choices for users to choose and then our system uses the chosen method for the obtaining of the classified result of the input text. There are two improved algorithms, one is k-Bayes using hierarchy conception based on NB method in machine learning field and another one adds attention layer to the convolutional neural network in deep learning field. Through experiments, our results showed that improved classification algorithms had better accuracy than based algorithms and our system is useful for making classifying news texts more reasonably and effectively.
机译:在本文中,我们设计了一种自动中文文本分类系统,旨在实施用于分类新闻文本的系统。我们提出了两个改进的分类算法作为用户选择的两个不同选择,然后我们的系统使用所选方法来获取输入文本的分类结果。有两个改进的算法,一个是基于机器学习领域的NB方法的层次结构的k-bayes,另一个是深度学习领域的卷积神经网络中的注意层。通过实验,我们的结果表明,改进的分类算法比基于算法更好的准确性,我们的系统对于更合理且有效地制作分类新闻文本是有用的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号